AbstractUnwanted power line interference is one of the most common problems in electroencephalographic recording. This paper examines how the use of active electrodes together with a driven-right-leg circuit can significantly improve interference reduction, even when the same electrode is used for common and reference which is attractive because it saves an electrode. General conclusions about the active electrodes and the driven-right-leg circuits were obtained thanks to a prototype that uses the same electrode for both common and reference. Measurements were performed both on a subject and on an electrical equivalent model. KeywordsActive electrode, bioelectric recording, driven- right-leg, reference electrode. I. INTRODUCTION Unwanted power line interference in electroencephalographic (EEG) recording is one of the most recurrent problems in hospitals. Capacitive coupling between line, ground and cables is one of the major sources of line interference [1]. Although some papers have previously been written on active electrodes (e.g. [2], [3] and [4]), and on driven-right- leg circuits (e.g. [5]), none of them seem to have published figures on quantitative tests on subjects. This paper examines how the use of active electrodes together with a driven-right-leg circuit can significantly improve interference reduction. Quantitative information about parasitic voltage on measurement and common electrodes was also deduced from measurement and can be generalized to all kinds of EEGs. The measurements were performed both on an electrical model and on a subject. In a traditional EEG, the common electrode is connected to the differential amplifiers common. The differential amplifiers amplify the signals of the measurement electrodes with respect to the reference electrode. Using two different electrodes for reference and common helps to reduce common-mode interference using the well known three-op- amps instrumentation amplifier configuration. Our prototype uses the same electrode for common and reference which is attractive because it saves an electrode. This is particularly important for application with a low number of electrodes (typically 8 for a portable holter EEG). Our approach was to build a two-channel battery powered prototype able to record an EEG signal simultaneously with and without an active electrode and able to switch from the driven-right-leg circuit to the traditional common electrode circuit. The prototype enables the comparison of those systems and the benefits from active electrode and driven-right-leg circuits were quantified. Reduction of Power Line Interference using Active Electrodes and a Driven-Right-Leg Circuit in Electroencephalographic Recording with a Minimum Number of Electrodes A. Nonclercq, P. Mathys Microelectronics and Electronics (MiEL) Department, UniversitØ Libre de Bruxelles, Brussels, Belgium This work has been funded by the FRIA. The results are in good agreement with theory and simulations. II. THEORETICAL APPROACH AND SIMULATIONS The prototype is presented in Fig. 1. It is composed of two measurement channels. The only difference between the two channels is the location of the pre-amplifier : while the pre-amplifier of the second channel is traditionally placed next to the amplifier (passive electrode), the pre-amplifier of the first is placed next to the electrode (active electrode). This way, it is possible to record simultaneously with and without an active electrode. It is also possible to switch the common electrode from the amplifiers common to the driven-right-leg circuit. Stray capacitances produce interference in the measurement and common wires that are nearly current sources because the impedances of the stray capacitances are much larger that those of the electrodes and the body. In a system without a driven-right-leg circuit (switch connected on common) the parasitic currents are lower in the measurement electrodes (Im1 and Im2) than in the common electrode (Ic) because the parasitic currents are nearly stopped by the pre-amplifiers high impedance and can only flow through the subject. So, a large parasitic current flows through the common electrode impedance and is the main cause of common-mode voltage. The driven-right-leg circuit drives the subject to the common average voltage of the input signals, reducing the Fig. 1. EEG system. 0-7803-8439-3/04/$20.00©2004 IEEE 2247 Proceedings of the 26th Annual International Conference of the IEEE EMBS San Francisco, CA, USA • September 1-5, 2004